skills/lelborn/skills/doc-check/Gen Agent Trust Hub

doc-check

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from the project's source code and .docs/ tree, which provides an attack surface for indirect prompt injection. Maliciously crafted content in the documentation could attempt to influence the auditor's summary or the fixes it suggests.
  • Ingestion points: Reads all files within the .docs/ directory and the authoritative source code tree as specified in SKILL.md.
  • Boundary markers: The instructions do not define specific delimiters to wrap external data or provide warnings to ignore instructions embedded within the files being audited.
  • Capability inventory: The skill performs read operations across the repository and has the capability to modify files (redact secrets, repair links) or delegate tasks to other skills (doc-it) when requested by the user.
  • Sanitization: The instructions lack specific requirements for sanitizing or escaping the content read from files before it is included in the generated report.
  • [DATA_EXPOSURE]: The skill proactively searches for committed secrets, API keys, and PII within documentation. This is an intended security feature designed to help users identify and redact sensitive information before publication, rather than an attempt at unauthorized data exfiltration.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 10:37 AM